Construction and Building Materials 55 (2014) 462469

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Bio-bricks: Biologically cemented sandstone bricks

D. Bernardi a, J.T. DeJong b,, B.M. Montoya c, B.C. Martinez d
a
ENGEO Incorporated, Walnut Creek, CA, USA
b
Department of Civil and Environmental Engineering, University of California, Davis, CA, USA
c
Department of Civil, Construction, and Environmental Engineering, North Carolina State University, Raleigh, NC, USA
d
Geosyntec Consultants, Oakland, CA, USA

h i g h l i g h t s

 A novel technique to manufacture bio-bricks using a biologically mediated natural cementation process is presented.
 Results show that bio-bricks can have compressive strengths up to 2 MPa.
 P-wave velocity measurements show bio-brick stiffness to be relatively uniform and high.
 Bio-bricks are comparable to bricks prepared with the more conventional cement and hydraulic lime additives.

a r t i c l e i n f o a b s t r a c t

Article history: The cementation of sand into sandstone through microbial activity is a novel technology with a wide
Received 24 September 2013 range of possible applications. The cementation process involves the introduction of bacteria and nutri-
Received in revised form 9 January 2014 ents to sand, and through bacterial processes calcite precipitation binds particles together, ultimately cre-
Accepted 11 January 2014
ating a sandstone material. This technology could provide a new, more sustainable building material in
Available online 14 February 2014
the form of bio-bricks. This paper describes the treatment technique as well as results from testing after
brick manufacturing. Bricks were tested to determine compression (p-wave) wave velocity, unconned
Keywords:
compression strength, and calcite concentration. P-wave velocity, stiffness, strength, and calcite content
Bricks
Calcite precipitation
of bio-bricks all increase with further treatment of bacteria and cementation media. Results show that
Microbially induced calcite precipitation bio-bricks can have strengths ranging from 1 MPa to 2 MPa. Bio-bricks are comparable in terms of stress
Compression strength and stiffness to bricks prepared with the more conventional cement and hydraulic lime additives.
Stiffness 2014 Elsevier Ltd. All rights reserved.

1. Introduction
The global use of resources and emphasis on sustainable infra-
structure is a growing societal issue civil engineers must address
[21]. The international population is growing at an unprecedented
rate, and in response, civil infrastructure must expand and be reha-
bilitated in a sustainable manner. The demand on natural resources
is far greater than the supply in both developed and developing coun-
tries [1]. Sustainable development must consider the energy and
material ows through the construction, maintenance, dismantling,
and material disposal related to a project [28]. Meeting the societal
demands with locally available resources and minimal material
and energy promote a sustainable approach to development.
Biological processes have been harnessed for a multitude of
engineering applications [9,10], DeJong et al. [11]. Bio-geochemical
Corresponding author. Address: Department of Civil and Environmental Engi-
neering, University of California, One Shields Ave., Davis, CA 95616, USA. Tel.: +1
530 754 8995.
E-mail address: jdejong@ucdavis.edu (J.T. DeJong).
processes that induce mineral precipitation have been utilized for
many applications, including improving the strength and stiffness
of soil [8,32,20] as an alternative to traditional chemical grouting
which can be environmentally hazardous [17]. Microbially induced
calcite precipitation (MICP) can be used for a variety of other appli-
cations including environmental remediation [13], improved dura-
bility and remediation of concrete [25,7], calcium removal in
wastewater [15], and carbon sequestration [26].
Although various forms of MICP are available using different
bacterial and precursors, the form of MICP treatment used for this
research utilized natural soil bacteria to metabolize urea, increas-
ing the pH of the pore water, promoting mineral precipitation. Ure-
olytic bacteria are prevalent in natural soils; they increase the
alkalinity of the soil by hydrolyzing the urea to produce ammonia
and carbon dioxide. This induces calcite precipitation primarily at
particleparticle contacts, which increases the strength and stiff-
ness of the sand. The amount of calcite cementation is proportional
to the concentrations of chemicals supplied (e.g. urea and calcium)
and the number of treatments performed. The reaction network for
the net urea hydrolysis reaction and formation of calcite is:

http://dx.doi.org/10.1016/j.conbuildmat.2014.01.019
0950-0618/ 2014 Elsevier Ltd. All rights reserved.
 D. Bernardi et al. / Construction and Building Materials 55 (2014) 462469
NH2  CO  NH2 3H2 O ! 2NH4 HCO3 OH
Ca2 HCO3 OH ! CaCO3 H2 O
Current methods for brick manufacturing vary widely, but most
methods include high energy processes of compression under high
stresses and/or baking at high temperatures. The most common
method of brick manufacturing is by ring clay at high tempera-
tures. Red clay bricks are typically placed in wood molds and dried
in the sun for 23 days and then baked in the oven for 24 h at tem-
peratures up to 1200 C [6]. Engineering properties and physical
characteristics differ between red clay bricks primarily due to the
clay source and ring temperature. For example, Lower Oxford Clay
based bricks have a 28 day unred strength of 3.5 MPa and a red
strength above 20 MPa [22] (and other references in Table 1). Alter-
natives to red clay bricks include sandlime bricks, which are man-
ufactured using water, sand, and lime mixed together, compacted
together at a pressure of 20 MPa and then autoclaved for up to 9 h
at temperatures of up to 190 C (Fang et al. [12]). Another method
uses clay in addition to lime, cement, and a manufacturing byprod-
uct such as ground granulated blast furnace slag. The bricks are
cured at room temperature, yielding strengths between 2.7 and
5 MPa [23]. Other methods of manufacturing earth-based building
materials consists of adobe, cob, rammed earth, and compressed
earth bricks [29,24]. The range of strengths and modulii for these
and other bricks are summarized in Table 1.
This paper summarizes a research program undertaken to
develop a natural, bio-mediated process for the manufacturing of
bio-bricks (Bernardi [4]). The materials, treatment methods, and
measurement techniques are presented rst. Bricks produced
using the novel technique are compared against cement and lime
treated bricks. Results assessing the treatment uniformity within
individual bricks, as measured and indicated by shear and com-
pression wave velocity, are presented. The correlation between
cementation level (precipitated calcite concentration) and wave
velocities is then investigated. A comparison of brick strength
between the three brick types is presented, followed by correla-
tions between compressive strength and velocity measurements.
2. Materials and test methods
2.1. Soil
The sand used for production of all bricks was silica rich #1 masonry sand, quar-
ried in Chico, California. This sand was used because it is moderately graded, locally
produced, and available in large quantities. Salient sand characteristics are pre-
sented in Table 2.
2.2. Bacteria and growth conditions
The soil bacterium utilized in this study was Sporosarcina paseurii (ATCC 11859).
Cultures were grown in an Ammonium-Yeast Extract media (ATCC 1376) as
described in Mortensen and DeJong [20] (0.13 M Tris Buffer, 10 g of (NH4)2SO4,
and 20 g of yeast extract per liter of deionized water). The bacteria were inoculated
in the growth media and incubated aerobically in a 30 C water bath shaken at
200 rpm for approximately 24 h. Bacteria were incubated until samples obtained
an optical density near 1.0 using a spectrophotometer (600 nm wavelength). The
Table 1
Strength of other bricks and materials.
Bricks Strength (Mpa) Elastic modulus (MPa)
Autoclaved bricks 20
Red clay bricks >20
Compressed earth block 0.73.1 200
Rammed earth 0.751.5 72102
Adobe 1.21.8 100300
Sandstone 70 45,000
Limestone 1070
[12,14,5,18,30,3,27,16].
463
Table 2
Sand characteristics.
Material D50 (mm) Cu Cc Gs emin emax Mineralogy
#1 Masonry sand 0.42 2.6 1.2 2.6 0.5 0.8 Quartz
sand was inoculated with the bacteria by percolating the bacterial solution through
the sand top-down, which was retained for 4 h in the soil before treatments with
cementation media began.
2.3. Cementation media
A urea-calcium medium was used to drive calcite precipitation. The cementa-
tion media consisted of urea (200 mM), calcium chloride (100 mM), and nutrient
broth (0.5 g/L). The nutrient broth, which contains beef extract and peptone, was
used to enable bacteria reproduction within the brick mold.
2.4. Brick mold
Three identical brick molds were fabricated from PVC plastic, with each mold
containing ve bricks with dimensions of 91 mm by 58 mm by 200 mm (similar
dimensions as standard red clay bricks, Fig. 1, Bernardi [4]). The mold is assembled
with screws and silicone sealant with drain holes at the mold base to enable uid to
percolate through. The mold base enables saturation of the mold during bacterial
treatment and relatively unobstructed ow during cementation treatment. Three
plastic screens with different opening sizes (3.360 mm, 0.711 mm, and 0.178 mm)
were placed at the mold base to prevent soil loss during treatment. The sand is then
placed, three additional screens were placed on top of the sand, and a low conning
stress (10 kPa) applied with a rubber band. Coarse gravel is placed on top to pre-
vent erosion of the sand when the treatment solution is added.
2.5. Preparation and treatment programs
2.5.1. MICP treatment method
The bacterial solution was added to the sand by percolation (i.e. unrestrained
ushing of uid from top to bottom). The treatment method implemented was
selected in order to ensure bacteria attachment at particle contacts within the per-
meable sand matrix. Efuent consisting of the bacterial solution was cycled through
the sand two additional times to improve bacteria attachment throughout the sam-
ple, and during the second cycle the mold was sealed to create fully saturated con-
ditions. Treatment media was added to the sand by percolation. Three brick molds
were treated for different amounts of time. Since the bio-bricks were going to be
compared to lime and cement treated bricks that were cured for up to 28 days,
an equivalent treatment time was devised. Treatments ranged from 1 to 5 times
per day, depending on permeability reduction from the treatment, so an average
of 3 treatments per day was dened as equivalent to one day of curing of conven-
tional bricks. The molds were treated at 7 days (21 treatments), 14 days (42 treat-
ments), and 28 days (84 treatments). A 12 h retention was usually allowed
overnight before treatment started the following day. All treatments contained
the cementation media (Table 2). pH readings were made of the inuent solution
and the immediate efuent of each brick with the use of pH strips (displaying pH
in the range of 6.59.0). Occasionally excess calcite precipitation on the injection
face of the bricks reduced permeability sufciently that the mold was partially dis-
assembled and the screens cleaned. Once the required treatments were completed,
two pore volumes of deionised water with 50 mM sodium chloride were percolated
through to rinse excess chemicals from the pore space. The mold was then disas-
sembled and the bricks were oven dried overnight in a 77 C oven. The brick dimen-
sions were measured and the mold was weighed again to estimate changes in dry
density and void ratio.
2.5.2. Lime treatment method
Lime bricks were prepared by combining dry sand with varied volumes of
hydraulic lime. The evaluated percents of hydraulic lime to sand by volume were
20%, 25%, 30%, 40%, and 50% for each set of ve bricks (these correspond to percent-
ages by weight of about 10.1%, 12.7%, 15%, 20.9%, and 26.7%). These mixtures brack-
et the strengths that were expected from the bio-bricks (2 MPa) and were selected
in part from manufacturer recommendations. The lime used was from the manufac-
turer St. Austier and is a natural hydraulic lime (NHL5) with approximately 2030%
clay included as the silica source. The lime and the sand were measured, dry mixed,
and then water was added until proper workability was achieved (250 mL of
water per brick). The sandlime mixture was placed in approximately 2.54 cm lifts
and tamped 50 times using the steel overburden stress tamper. The overburden
stress was then applied the same way as the bio-brick treatment. The three brick
molds were used to make batches of bricks to be tested at different curing times.
The bricks set for 2 days and then cured for 7, 14, and 28 days in a constant humid-
ity chamber (95% humidity, 13.3 C). After curing was complete, the brick molds
were disassembled and bricks dried for up to 2 days in a 77 C oven before testing.
464 D. Bernardi et al. / Construction and Building Materials 55 (2014) 462469

Overburden Stress
Rip Rap
Top Filter Screens

Brick
Divider
Bottom Filter Screens and Drain Holes

Fig. 1. Image of assembled brick mold (left image with one side removed for display purposes).

2.5.3. Cement treatment method 2.6. Measurement methods

Cement bricks were prepared by combining dry sand to various volumes of
type II/V cement according to ASTM C150. The cement was added in quantities
of 5%, 10%, 15%, 20%, and 25% by volume for each respective brick in a batch
(corresponding to percents by weight of about 3.7%, 7.4%, 11.2%, 15.3%, and
19.3%). The cement amount selected was to bracket the expected strengths of
the bio-bricks. The cement and sand were measured and manually mixed
together dry and then water was added until a proper consistency for workability
was achieved (250 mL per brick). The mixture was then placed in approximately
2.54 cm lifts and tamped 50 times using the overburden stress tamper. An
overburden stress was applied. The bricks set for one day and then cured for 7,
14, and 28 days in a constant humidity chamber (95% humidity, 13.3 C). After
curing was completed, the bricks were dried for up to 2 days in a 77 C oven
before testing.
2.6.1. Compression wave velocity
Seismic wave measurements are effective in monitoring the incremental
cementation that occurs with MICP [8,9,31]. As cementation occurs the contacts
between particles become increasingly stiff, which results in an increase in bulk
stiffness and faster transmission of seismic waves. Compression seismic waves
(p-waves) were used post-treatment to assess cementation uniformity along the
brick length for the bio-bricks, hydraulic lime, and cement bricks. The equipment,
settings, and techniques used are specied by Weil et al. [31]. A pair of 0.5 MHz
Panametrics V101-RB ultrasonic transducers were powered by an HP 33120A Func-
tion Generator and received by a Fluke PM3384 oscilloscope. The transmitted signal
was a +10 V, 60 kHz square wave and the received signal was subjected to a 1 kHz
high pass lter to remove background noise and a 30 dB gain. Five equidistant loca-
tions along the brick lengths were measured. All of the p-wave measurements were
performed across the brick width (91 mm). Vacuum grease was used to provide
coupling between the transducer faces and brick surfaces.

7 Day 2.6.2. Unconned compressive strength

0 Unconned compression testing was performed on all bricks in accordance with
Distance From Top

(a) ASTM C67-07a [2]. Tests were performed on a GeoTac Sigma-1 Triax machine with
5
of Brick (cm)

a 4.5 Mg (10 kip) load frame. Bricks were cut in thirds instead of halves, as recom-
mended in the ASTM standard, to better assess brick spatial variability. All bricks
Brick 1
were cut with a wet masonry saw to minimize disturbance. The dimensions of both
10 Brick 2
faces of each brick section undergoing loading were taken, as well as the average
Brick 3
depth of the brick. All surfaces of the testing apparatus were cleaned and the spec-
15 Brick 4
imens were placed without sulfur or gypsum capping because surfaces of the bricks
Brick 5
were at, smooth, and parallel. The load was seated on the specimen and testing
20 initiated. The test was strain controlled with a rate of 2% strain/min (0.045 cm/
0 1000 2000 3000 4000 min). Tests were performed until failure and the maximum stress was recorded [2].
P-wave Velocity (m/s)
2.6.3. Calcite measurements
14 Day Calcite concentration measurements were performed following the methods
0 described in Mortensen and DeJong [20] and Martinez et al. [19]. Samples were
Distance From Top

obtained from the middle of each brick section after compression testing. The dry
(b) weight of the sample was taken, then acid washed with 5 M HCl, and dry weights
5
of Brick (cm)

obtained again after acid washing to determine the amount of cementation that
Brick 6 occurred. The sand used does contain a small percentage of nes which is lost when
10 Brick 7 undergoing this treatment, so the calcite test was also performed with un-treated
Brick 8 sand to determine the average loss of nes and any calcite existing in the soil. This
15 Brick 9 was determined to be approximately 0.88% of the sample, so the values presented
Brick 10 are adjusted for the loss in nes.
20
0 1000 2000 3000 4000
3. Results and discussion
P-wave Velocity (m/s)

28 Day 3.1. Small-strain properties and uniformity

0
Distance From Top

The compression wave (p-wave) velocities along the length of

(c) the brick are plotted in Fig. 2 with the ranges, average, and
5
of Brick (cm)

Brick 11 coefcient of variability presented in Table 3. P-wave velocity for

10 Brick 12
Brick 13
Brick 14 Table 3
15
Brick 15 Minimum, maximum, average, and coefcient of variation for p-waves of bio-bricks.

20 Treatment time Average Vp Min Vp Max Vp COV

0 1000 2000 3000 4000 (Days) (m/s) (m/s) (m/s) (%)
P-wave Velocity (m/s) 7 1560.8 1383.6 2591.5 6.0
14 2269.9 2008.0 2591.5 6.9
Fig. 2. P-wave velocities of bio-bricks at (a) 7 day, (b) 14 day, and (c) 28 day 28 3062.7 2723.4 3576.5 6.6
treatment.
 D. Bernardi et al. / Construction and Building Materials 55 (2014) 462469 465

0 brick (range of 177 m/s), but the top portion exhibits the highest
(a) compression wave velocity with a slight decrease toward the brick
Distance From Top

5 base. The trend of p-wave for the 14 day bio-brick batch have a
of Brick (cm)

20% Lime range of 262 m/s and is slightly higher at the top of the brick than
10 25% Lime at the bottom of the brick, indicating more cementation occurring
30% Lime
in the top of the brick and gradually decreasing throughout the
40% Lime
15 brick length (Fig. 6b). The p-wave velocities for the 28 day bio-
50% Lime
brick batch have a range of 339 m/s. The top of the brick possesses
20
0 1000 2000 3000 4000 a high p-wave velocity with a decrease toward the upper third to
P-wave Velocity (m/s) the brick center. P-wave velocity then increases again towards
the brick base, which results in a p-wave velocity being compara-
0 ble to the top of the brick. The biased cementation at the brick top
Distance From Top

(b) is most likely due to a majority of the bacteria being present at the
top of the brick (since bacteria were initially ushed top-down).
of Brick (cm)

5 5% Cement
10% Cement P-wave velocities of lime and cement bricks for the range of
10 15% Cement treatment levels and time periods showed consistent spatial uni-
20% Cement formity and increasing velocity as expected. Representative results
25% Cement
15

20 Table 4
0 1000 2000 3000 4000
Minimum, maximum, average, and coefcient of variation for strength values of bio-
P-wave Velocity (m/s) bricks.

Fig. 3. P-wave velocities of lime and cement treated bricks after 28 day curing time. Treatment time Average Min strength Max strength COV
(Days) strength (kPa) (kPa) (kPa) (%)
7 69.1 43.3 121.7 37.0
7 Day 14 441.3 115.5 934.7 48.7
200
28 1645.4 910.9 2286.6 26.4
(a)
150
Stress (kPa)

100
Table 5
Minimum, maximum, average, and coefcient of variation for E50 modulus values of
50
bio-bricks.

0 Treatment time Average E50 Min E50 Max E50 COV

0 2 4 6 (Days) (kPa) (kPa) (kPa) (%)
Strain (%) 7 5782.4 3140.6 13868.5 49.2
14 15019.5 4525.0 29964.7 47.4
14 Day 28 83915.5 27433.8 131398.3 36.5
1000
(b)
750
Stress (kPa)

500 40% Lime

2000
(a)
250
Stress (kPa)

0 1000
0 2 4 6
Strain (%)

28 Day 0
2000 0 2 4 6
(c) Strain (%)
Stress (kPa)

20% Cement
1000 2000
(b)
Stress (kPa)

0 1000
0 2 4 6
Strain (%)

Fig. 4. Representative stress strain curves of bio bricks tested in unconned

0
compression. 0 2 4 6
Strain (%)
bio-bricks exceeds 3500 m/s with an average coefcient of varia-
tion within a single brick of 6.5%. The trend of the p-wave velocity Fig. 5. 28 Day stress strain curves of 40% line and 20% cement bricks tested in
for the 7 day bio-brick indicates uniformity within the center of the unconned compression.
466 D. Bernardi et al. / Construction and Building Materials 55 (2014) 462469

after 28 days of curing for all of the lime and cement treated bricks
are shown in Fig. 3 (with further details in Bernardi [4]). An
increase in curing time resulted in very little change in p-wave
velocities; on average p-wave velocities increased by 250 m/s for
both lime and cement bricks. The average coefcient of variability
within a single brick for the lime bricks was 12.8% while the
cement bricks were 15.8%.
3.2. Compressive strength and secant modulus
Stress strain curves of bio-bricks were obtained from compres-
sion testing on ve bricks at each treatment interval, with repre-
sentative stress strain curves presented in Fig. 4. The peak
strength of the stress strain curve and the modulus (computed as
a secant elastic modulus based on the strain required to mobilize

Table 6
Constants for statistical ts of strength and modulus plotted with p-wave velocity.

Admixture X-axis Y-axis A B C R2

MICP P-wave Strength 0 0.0016 2.45 0.75
P-wave E50 423 0.0018 5.61 0.38
P-wave Calcite content 5 0.0004 1.55 0.83
Calcite content Strength 5265 0.0232 8.49 0.75
Calcite content E50 6527 0.2052 8.1 0.38
Lime P-wave Strength 140 0.0010 4.69 0.65
P-wave E50 16,010 0.0006 9.54 0.34
P-wave Percent mass 50 0.0003 3.84 0.81
Percent mass Strength 0 0.1038 3.84 0.75
Percent mass E50 0 0.0957 7.78 0.70
Cement P-wave Strength 70 0.0011 4.14 0.83
P-wave E50 1570 0.0014 7.00 0.63
P-wave Percent mass 9 0.0004 2.05 0.97
Percent mass Strength 400 0.1144 5.90 0.87
Percent mass E50 24,345 0.0978 10 0.53

Bio-Bricks
(a) 3000 200000
E 50 Modulus (kPa)

7 Day 7 Day R2 = 0.38

R2 = 0.76
Strength (kPa)

14 Day 150000 14 Day

2000 28 Day 28 Day
100000
1000
50000

0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000
P-wave Velocity (m/s) P-wave Velocity (m/s)

Lime Bricks
(b) 1000 100000
E 50 Modulus (kPa)

7 Day R2 = 0.34
Strength (kPa)

7 Day
750 14 Day 75000 14 Day
28 Day 28 Day
500 50000
R2 = 0.65
250 25000

0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000
P-wave Velocity (m/s) P-wave Velocity (m/s)

Cement Bricks
(c) 4000 250000
E 50 Modulus (kPa)

R2 = 0.83 R2 = 0.63
200000
Strength (kPa)

7 Day 7 Day
3000 14 Day 14 Day
28 Day 150000 28 Day
2000
100000
1000
50000

0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000
P-wave Velocity (m/s) P-wave Velocity (m/s)

Fig. 6. 28 Day strength and E50 modulus versus p-wave velocity for (a) bio-bricks, (b) lime bricks, and (c) cement bricks.
 D. Bernardi et al. / Construction and Building Materials 55 (2014) 462469 467

50% of the peak strength, E50) was computed for each brick section while the tangent secant modulus, E50, of bio-bricks only reach
tested. The average and range of strengths and modulii as well as up to 130 MPa.
the coefcient of variation are presented in Tables 4 and 5, respec-
tively. The 7-day treatments of bio-bricks have a relatively low
3.3. Relation to calcite/admixture concentrations
strength and modulus exhibiting a ductile failure (Fig. 4a). After
14 days of treatment the bio-bricks increase in both strength and
Results showing percent of calcite along the length of the bio-
modulus with a transition between ductile and brittle behavior
bricks are presented in Fig. 7. Trends of calcite concentration along
that can be seen between the top and middle/bottom segments
the length of the bricks seem to be counterintuitive at rst, indi-
(Fig. 4b). Upon further treatment to 28 days, the strength and mod-
cated a smaller amount of calcite at the top of the brick where
ulus continue to increase throughout the brick with the top of the
the higher strength occurs (Fig. 7a). The 14 day bio-bricks are
brick being stronger than both the middle and bottom. The bottom
slightly more varied than the 7 day, with a couple of bricks con-
of the brick becomes stronger than the middle portion of the brick
taining higher calcite at the top of the brick and gradually decreas-
at this treatment level, which coincides with the p-wave data
ing along the brick. Bricks 9 and 10 of Fig. 7b are exceptions
(Fig. 3c). All brick segments exhibit brittle failure after 28 days of
containing a low calcite percentage at the top relative to what is
treatment (Fig. 4c). Overall, strengths and modulii both increase
expected to occur. For the 28 day batch in Fig. 7c the calcite con-
with continued treatment. The average strength increases from
tent variability is fairly narrow at the top of the brick, widens in
69 kPa to 1600 kPa, while the average modulus increases from
the middle, and then narrows again towards the bottom of the
5782 kPa to over 83,900 kPa from 7 to 28 days of treatment.
brick. This data is consistent with the p-wave and strength data
Representative stress strain curves for 40% lime and 20% cement
collected for the 28 day bio-bricks where the values of all these
bricks cured at 28 days are plotted in Fig. 5 for comparison. Lime
properties are higher in the top and bottom portions of the brick
and cement bricks show a general trend of the top and middle
than in the center. The trends of calcite, particularly observed in
brick segments having a higher strength and modulus than the
the 7 and 14 day data, with increasing calcite along the brick
bottom brick segment. Lime bricks exhibit ductile failure exclu-
length that are not congruent with the p-wave data are attributed
sively, while cement exhibits ductile failure at a low admixture
to a measurement error associated with the calcite measurement
percent (10% and lower) while changing to brittle failure above
technique. As mentioned previously, 0.88% of the sample consists
10% concentration. Strength and modulii for lime and cement sys-
of ne soil particles that could be washed out during the calcite
tematically increased with increase in admixture (Bernardi [4]).
Strength and modulii for lime bricks remain low relative to the
strength gain of both bio-bricks and cement bricks. The lime bricks 7 Day
tested at 50% lime content had a maximum strength of 0.98 MPa 0
Distance From Top

and a maximum secant modulus of 544 MPa, which is less than Brick 1
of Brick (cm)

that claimed by the manufacturer (compressive strength of 5 Brick 2

2.2 MPa, elastic tangent modulus of 10,800 MPa). The source of Brick 3

the difference was investigated with additional tests, but could 10 Brick 4
Brick 5
not be identied. Compression strengths and modulii for cement
15
bricks tend to increase more rapidly with increasing percent of
admixture.
(a)
20
Strengths and modulii of bio, lime, and cement bricks are plot- 0 5 10 15 20
ted versus p-wave velocity in Fig. 6. A three constant single expo- Calcite Content (%)
nential growth equation as shown in the plots was tted to each
data set: 14 Day
0
Distance From Top

y A eBxC 1 Brick 6
of Brick (cm)

5 Brick 7
Brick 8
The constants for each t are presented in Table 6. Constant A is
chosen to anchor a p-wave velocity between 150 and 300 m/s based 10 Brick 9
Brick 10
off of values from previous research on dry sand with 100 kPa of
15
connement. The R2 values for strengths versus p-wave are above (b)
0.65 while the R2 for modulus versus p-wave are higher than
20
0.34. The plots of modulus generally have a higher data spread, 0 5 10 15 20
resulting in lower R2. In all cases, brick strength and modulus gen- Calcite Content (%)
erally increase with an increase in p-wave velocity.
Bio-bricks range in strengths from 1.0 to 2.2 MPa, lime reaches
28 Day
as high as 1.0 MPa, and cement has a maximum strength of 0
Distance From Top

2.5 MPa. Other brick materials made of earth and sustainable Brick 11
of Brick (cm)

material such as adobe and rammed earth (Table 1) have strengths 5 Brick 12
ranging from 0.7 MPa to 3.1 MPa. A more conventional building Brick 13
material such as red red clay brick has strengths above 20 MPa, 10 Brick 14
Brick 15
much higher than bio-bricks. The p-wave velocities of bio-bricks
are comparable to sandstone. Natural sandstone and limestone 15
have an unconned compressive strength of up to 70 MPa, which (c)
is much higher than any material tested herein. A direct compari- 20
0 5 10 15 20
son of the tangent secant modulus reported herein cannot be
Calcite Content (%)
directly compared to the elastic modulus values in Table 1 as these
values are for small strain elastic modulus. For example, the mate- Fig. 7. Percent calcite for bio-brick for (a) 7 day, (b) 14 day, and (c) 28 day
rials in Table 1 can have an elastic modulii up to 300,000 MPa treatment times.
468 D. Bernardi et al. / Construction and Building Materials 55 (2014) 462469

measurement. These same ne particles could have rst migrated 1

downwards inside the bricks during treatment, resulting in higher emax (a)
0.8
nes content near the brick base than at the top. When the calcite

Void Ratio
measurement is then performed and the average value of 0.88% 0.6 emin
subtracted to correct for the average nes content this likely
results in an overcorrection for nes near the specimen top (result- 0.4 7 Day
14 Day
ing in too low of a calcite estimate) and an under correction near
0.2 28 Day
the base (resulting in too high of a calcite estimate). Unfortunately
the experimental program did not allow this issue to be explored 0
0 1000 2000 3000 4000
further.
The relationship between calcite content, p-wave velocity, P-wave Velocity (m/s)
strength, and modulus are presented in Fig. 8. A trendline using
Eq. (1) above was t to all three relationships. In general, p-wave 2.25

Dry Density (Mg/m 3)

velocity, strength, and modulus increase with calcite content with 7 Day (b)
R2 values of 0.83, 0.75, and 0.38, respectively. The p-wave velocity 2 14 Day
was again anchored to the range of 150300 m/s as explained pre- 28 Day
viously. The plot of calcite content versus strength was best t d max
1.75
when anchored to a calcite content of approximately 3.4%, which
was then used for the graph of modulus. Similar trends were 1.5 d min
observed between the lime and cement admixture concentration
versus p-wave velocity, strength, and modulus (Bernardi [4]). Like 1.25
0 1000 2000 3000 4000
the bio-bricks, plots of percent of mass of admixture with p-wave
velocity and strength yield a higher R2 values than when plotted P-wave Velocity (m/s)
with modulii.
Fig. 9. Final dry density versus p-wave velocity for bio-bricks.

3.4. Void ratio and dry density

dimensions and mass of the full dry bio-bricks after treatment
The average nal void ratio and nal dry density after treatment and the average p-wave velocities for an entire brick. The average
to the designated time increments are presented versus p-wave initial void ratio for each bio-bricks was 0.51, and the initial dry
velocity in Fig. 9. The data plotted was calculated using the density was 1.72 Mg/m3. As evident, the void ratio following MICP
treatment decreases to about 0.33, a value signicantly less than
20 the minimum void ratio for the untreated sand. The total dry den-
Calcite Content (%)

7 Day R2 = 0.83 sity of the bricks increases with treatment up to a value of about
15 14 Day 1.9 Mg/m3. The relatively small increase in density compared to
28 Day
the reduction in void ratio is attributed to the low dry density of
10
precipitated calcite relative to that of the silica sand particles.
5
(a)
4. Conclusions
0
0 1000 2000 3000 4000
P-wave Velocity (m/s) The novel MICP bio-brick manufacturing technique developed
herein can produce compressive strengths that exceed 2.0 MPa. Re-
3000 sults from 7, 14, and 28 day treatment schemes indicate that the
7 Day R2 = 0.75 strength and stiffness is scalable, enabling customization to project
Strength (%)

14 Day and site specic requirements. The strengths obtained after these
2000 28 Day different treatment durations followed trends similar to sand-based
bricks treated with lime (27% by weight) and cement (1218% by
1000 weight), though treatment with cement increased strength more
(b) consistently with respect to concentration of admixture.
The stressstrain behavior became more brittle with increasing
0
0 5 10 15 20 cementation level, eventually exhibiting behavior after 28 day
Calcite Content (%) treatment consistent with that for lime and cement admixture
bricks. Secant elastic modulus corresponding to 50% mobilized
150000 strength, capturing the stiffness upon initial loading, increased
from 5300 kPa after 7 days of treatment up to 83,900 kPa after
E50 Modulus (kPa)

7 Day R2 = 0.38
14 Day 28 days of treatment. The secant elastic modulus also increased
100000 28 Day with cement admixture concentration and to a lesser extent when
lime was used.
50000 Compression wave, or p-wave, velocity is effective in mapping
(c) the extent and uniformity of calcite cementation along the length
of the bricks, with p-wave values after 28 days of treatment
0 exceeding 3500 m/s and being comparable to sandstone and lime-
0 5 10 15 20
stone rock. Variability within individual bio-bricks averaged about
Calcite Content (%)
6.5% in terms of coefcient of variation (COV), indicating high spec-
Fig. 8. Calcite content plotted with (a) p-wave velocity, (b) strength, and (c) E50 imen uniformity. In comparison, lime and cement treated bricks
modulus. had average COV values of 11.1% and 6.4%, respectively. P-wave
 D. Bernardi et al. / Construction and Building Materials 55 (2014) 462469
velocity also correlated directly with admixture concentration,
secant elastic modulus, and strength for MICP, lime, and cement
additives. Correlations developed generally followed exponential
growth trends, with R2 values greater than 0.6. For the bio-bricks
specically, estimates of calcite, secant elastic modulus and
strength could be predicted with an accuracy of about 30%.
Overall this research has demonstrated the strong potential of
manufacturing sand-based bricks using a natural microbial pro-
cess. This is attractive given its simplicity, low embodied energy
in materials, and scalability. Further research is needed to obtain
a more accurate estimate on cost, durability, and net carbon foot-
print of MICP brick production relative to more conventional
methods.
Acknowledgements
Funding provided by the United States National Science Foun-
dation (#072746). Any opinions, ndings, and conclusions or rec-
ommendations expressed in this material are those of the
writer(s) and do not necessarily reect the views of the National
Science Foundation.
References
[1] American Society of Civil Engineers (ASCE). Policy statement 418: the role of
the engineer in sustainable development; 2004.
[2] ASTM. Standard test methods for sampling and testing brick and structural
clay tile. American Society for Testing and Materials (ASTM), Standard C67;
2007. p. 12.
[3] Baykasoglu A, Gullu H, Canakci H, Ozbakir L. Prediction of compressive and
tensile strength of limestone via genetic programming. Exp Syst Appl
2008:11123.
[4] Bernardi DJ. Biologically Cemented Sandstone Brick, MSCE Thesis, Department
of Civil and Environmental Engineering. University of California Davis; 2012. p.
53.
[5] Bui Q, Morel J. Assessing the anisotropy of rammed earth. Constr Build Mater
2009;23:300511.
[6] Custodio-Garcia E, Sebastian PJ, Campos-Alvarez J, Trevino-Palacios CG, Zarate
EA, Cordova QA, et al. Solar conduction heat transfer in red clay bricks. Sol
Energy Mater Sol Cell 2005;88:16978.
[7] de Muynck W, Debrouwer D, de Belie N, Verstraete W. Bacterial carbonate
precipitation improves the durability of cementitious materials. Cem Concr
Res 2008(38):100514.
[8] DeJong JT, Fritzges MB, Nusslein K. Microbially induced cementation to control
sand response to undrained shear. J Geotech Geoenviron Eng
2006;11(132):138192.
[9] DeJong JT, Mortensen BM, Martinez BC, Nelson DC. Bio-mediated soil
improvement. Ecol Eng 2010;36:197210.
[10] DeJong JT, Soga K, Banwart SA, Whalley WR, Ginn T, Nelson DC, et al. Soil
engineering in-vivo: harnessing natural biogeochemical systems for
469
sustainable, multi-functional engineering solutions. J R Soc Interface
2011;8:115.
[11] DeJong JT, Soga KS, Kavazanjian E, Burns S, van Paassen L, Al Qabany A.
Biogeochemical Processes and Geotechnical Applications: Progress,
Opportunities, and Challenges. Geotechnique 2013;63(4):287301.
[12] Fang Y, Kang Q, Wen Q, Dai P. Utilization of copper tailing for autoclaved sand
lime brick. Constr Build Mater 2011(25):86772.
[13] Fujita Y, Taylor JL, Gresham TLT, Delwiche ME, Colwell FS, McLing TL, et al.
Stimulation of microbial urea hydrolysis in groundwater to enhance calcite
precipitation. Environ Sci Technol 2008;8(42):302532.
[14] Hall M, Djerbib Y. Rammed earth sample production: context,
recommendations and consistency. Constr Build Mater 2004;18:2816.
[15] Hammes F, Seka A, Knijf S, Verstraete W. A novel approach to calcium
removal from calcium-rich industrial wastewater. Water Res 2002;37(3):
699704.
[16] Hendry AW, Sinha BP, Davies SR. Design of masonry structures. Taylor and
Francis e-Library; 2004.
[17] Karol RH. Chemical grouting and soil stabilization. New York (NY): Marcel
Dekker; 2003. p. 558.
[18] Lima S, Varum H, Sales A, Neto VF. Analysis of the mechanical properties of
compressed earth block masonry using the sugarcane bagasse ash. Constr
Build Mater 2012;35:82937.
[19] Martinez BC, DeJong JT, Ginn TR, Montoya BM, Barkouki TH, Hunt C, et al.
Experimental optimization of microbial-induced carbonate precipitation for
soil improvement. ASCE J Geotech Geoenviron Eng 2013;139(4):58798.
[20] Mortensen BM, DeJong JT. Strength and stiffness of MICP treated sand
subjected to various stress paths. In: Proceeding from ASCE geo-frontiers
2011 conference, Dallas, Texas; 2011.
[21] National Research Council (NRC). Geological and geotechnical engineering in
the new millennium: opportunities for research and technology
innovation. Washington (DC): The National Academies Press; 2006. p. 206.
[22] Oti JE, Kinuthia JM, Bai J. Engineering properties of unred clay masonry
bricks. Eng Geol 2009;107:1309.
[23] Oti JE, Kinuthia JM, Bai J. Unred clay bricks: from laboratory to industrial
production. Proc Inst Civ Eng 2009(ES4):22937.
[24] Pacheco-Torgal R, Jalali S. Earth construction: lessons from the past for future
eco-efcient construction. Constr Build Mater 2012;29:5129.
[25] Ramachandran SK, Ramakrishnan V, Bang SS. Remediation of concrete using
micro-organisms. ACI Mater J 2001;1(98):39.
[26] Renforth P, Manning DA, Lopez-Capel E. Carbonate precipitation in articial
soils as a sink for atmospheric carbon dioxide. Appl Geochem
2009;24(9):175764.
[27] Ruffolo RM, Shakoor A. Variability of unconner compressive strength in
relation to number of test samples. Eng Geol 2009;108:1623.
[28] Russ TH. Site planning and design handbook. The McGraw-Hill Companies,
Inc.; 2002. p. 443.
[29] Silveira D, Varum H, Costa A, Martins T, Pereira H, Almeida J. Mechanical
properties of adobe bricks in ancient constructions. Constr Build Mater
2012;28:3644.
[30] Tanacan L. Adobe construction: a case study in Turkey. Arch Sci Rev
2011;51.4:34959.
[31] Weil MH, DeJong JT, Martinez BC, Mortensen BM. Seismic and resistivity
measurements for real-time monitoring of microbially induced calcite
precipitation in sand. ASTM J Geotech Test 2011;35(2).
[32] Whifn VS, van Paassen LA, Harkes MP. Microbial carbonate precipitation as a
soil improvement technique. Geomicrobiol J 2007(24):41723.